IC Detection Test System Based on Embedded Vision
Date of Award
12-1-2023
Document Type
Thesis
Degree Name
Master of Science in Electronics Engineering
First Advisor
King Harold A. Recto, PhD
Abstract
One serious issue in the microcontroller manufacturing environment is the mixing of microcontroller unit (MCU) Integrated Circuit (IC) parts which leads to the wastage of materials, dissatisfied customers, and the implementation of non-value-adding activities to address the issues. More adverse effects include negative feedback from customers resulting in loss of confidence which affects the growth and development of the business. One of the root causes of the mixing of parts can be traced back to the final testing of the manufacturing back-end process when reusing unemptied standard JEDEC Matrix Trays (JMT) as the medium of good and bad units in the test handler. Currently in Microchip Technology Operations Philippines Corp (Mphil1), the leading test facility of 8-bit AVR® microcontrollers (MCUs) and 32-bit SAM MCUs, emptying and inspection of JMT is a manual process. This process is prone to human error considering the high-volume test run, small sizes of packages, and the color of the package having the same as the color of the JMT resulting in a high probability of reusing unemptied JMT. To eliminate the mixing of MCU parts issue, this study proposed an IC Detection Test System based on Embedded Vision to automate the inspection of the MCU Integrated Circuit (IC) on the JMT to determine if the JMT is empty or not. The test system puts forward the MCU IC -in-pocket detection by classifying MCU IC and background (neither MCU IC nor JMT) using a TinyML-based Convolutional Neural Network (CNN) that deployed in low-cost, low-power Embedded Vison OpenMV Cam sensor board. The test system also has an MCU-based Control Unit (CU) that automates the placement of OpenMV Cam board and JMT during the testing and communicates to OpenMV Cam for feedback on testing results. The IC Detection Test System achieved 97.33% accuracy in detecting IC on the JMT.
Recommended Citation
Pallones, Mark M., (2023). IC Detection Test System Based on Embedded Vision. Archīum.ATENEO.
https://archium.ateneo.edu/theses-dissertations/856
